Open rrryan2016 opened 4 years ago
I guess the problem may be that my backbone is not well trained, as it could extract detailed features as shown in the illustration pictures.
The pretrained on part of ImageNet is as below, and I am still working at it. | Baseline Backbone | My Backbone | |
---|---|---|---|
Train TOP1 Acc. | 83.792 | 72.572 | |
Train TOP5 Acc. | 94.957 | 89.900 | |
Val TOP1 Acc. | 82.760 | 74.800 | |
Val TOP5 Acc. | 94.580 | 90.760 |
I well-trained my backbone again, as below:
Baseline Baseline | My backbone | |
---|---|---|
Train TOP1 Acc. | 83.792 | 81.660 |
Train TOP5 Acc. | 94.957 | 93.940 |
Val TOP1 Acc. | 82.760 | 81.280 |
Val TOP5 Acc. | 94.580 | 94.100 |
However, the wacky problem still occurs,
Could anyone give me any advice? Thanks in advance.
Thanks for your great work and kind sharing.
I successfully reproduce the edge detection result, as
After NMS process, it turns like:
![2018](https://user-images.githubusercontent.com/24492366/90303338-d1f83780-dedf-11ea-987c-3fbe2337524e.png)
2018.jpeg
:2018.png
:However, when it switch the normal convolution to some special convolution in the same backbone(VGG16), the result is kinda wacky:
![2018](https://user-images.githubusercontent.com/24492366/90303394-39ae8280-dee0-11ea-9743-f1694e9596e1.png)
After same NMS process of edge thinning, it turns like more coarser in detail when compared with result above :![2018](https://user-images.githubusercontent.com/24492366/90303409-52b73380-dee0-11ea-88a2-4a91fad6cfb6.png)
Also, after evaluation, the
ODS-F
is 0.685891 andOIS-F
is 0.703985, both are much less than the paper, 0.788864 and 0.806692.Could any one please point out where is the potential problem lie in, cause I am really confused about it, and have no much experience in edge detection.
Thanks in advance !